The majority of older adults suffer multiple chronic conditions (MCCs): two-thirds of Medicare beneficiaries possess two or more MCCs, and 1 in 10 suffer 6 or more MCCs. As a result, older adults utilize 7 or more unique chronic medications annually. Thus, medication adherence, the extent to which a person's behavior coincides with medical or health advice, becomes particularly problematic in older adults with MCCs due to their greater exposure to multiple diseases and pharmacologic treatments. Poor medication adherence is implicated in clinically important and largely preventable health outcomes, including emergency department visits and hospitalizations, mortality, and medical costs. This study establishes the foundation needed to understand the management of two highly prevalent and costly conditions in older adults: chronic obstructive pulmonary disease (COPD) and depression. We focus on these two chronic, highly comorbid conditions because of their high prevalence, high costs, and their responsiveness to pharmacologic medication use and adherence. The objective of this study is to conduct research that informs our understanding of the cyclical relationships between depression and COPD-and the comparative effectiveness of treating each of these conditions-on important patient outcomes in older adults with COPD and depression. Using 2006-2011 Chronic Condition Data Warehouse (CCW) data, a nationally-representative Medicare administrative dataset of inpatient, outpatient, and pharmaceutical claims, we apply novel marginal structural modeling techniques to longitudinal analytic files. We will address the following three study aims: Aim 1) Test whether comorbid depression influences use of and adherence to necessary maintenance medications used to treat COPD; 2) Among Medicare beneficiaries with COPD and depression, test whether use of and adherence to antidepressant medications influences use of and adherence to necessary maintenance medications for COPD; and 3) Among Medicare beneficiaries with COPD and comorbid depression treated for depression, test whether antidepressant and/or COPD medication adherence reduces hospitalization, medical costs, and mortality. This study will produce the first longitudinal and nationally-representative estimates of depression incidence and treatment adherence in an older adult cohort diagnosed with COPD and depression, as well as improve our understanding of how depression treatment influences COPD maintenance medication use and adherence, and how COPD and depression treatment jointly influence important patient outcomes. The expected outcomes of this research are multiple, and include providing the knowledge needed to reduce preventable adverse outcomes, evaluate the effectiveness of depression treatment on important patient-relevant outcomes, and to demonstrate a sound methodological approach transferable to observational studies of other MCC combinations in Medicare and other available secondary databases.